split.c revision 1.15
1 /* $NetBSD: split.c,v 1.15 2003/06/26 23:06:52 bjh21 Exp $ */ 2 3 /* 4 * Copyright (c) 1987, 1993, 1994 5 * The Regents of the University of California. All rights reserved. 6 * 7 * Redistribution and use in source and binary forms, with or without 8 * modification, are permitted provided that the following conditions 9 * are met: 10 * 1. Redistributions of source code must retain the above copyright 11 * notice, this list of conditions and the following disclaimer. 12 * 2. Redistributions in binary form must reproduce the above copyright 13 * notice, this list of conditions and the following disclaimer in the 14 * documentation and/or other materials provided with the distribution. 15 * 3. All advertising materials mentioning features or use of this software 16 * must display the following acknowledgement: 17 * This product includes software developed by the University of 18 * California, Berkeley and its contributors. 19 * 4. Neither the name of the University nor the names of its contributors 20 * may be used to endorse or promote products derived from this software 21 * without specific prior written permission. 22 * 23 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 24 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 25 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 26 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 27 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 28 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 29 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 30 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 31 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 32 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 33 * SUCH DAMAGE. 34 */ 35 36 #include <sys/cdefs.h> 37 #ifndef lint 38 __COPYRIGHT("@(#) Copyright (c) 1987, 1993, 1994\n\ 39 The Regents of the University of California. All rights reserved.\n"); 40 #endif /* not lint */ 41 42 #ifndef lint 43 #if 0 44 static char sccsid[] = "@(#)split.c 8.3 (Berkeley) 4/25/94"; 45 #endif 46 __RCSID("$NetBSD: split.c,v 1.15 2003/06/26 23:06:52 bjh21 Exp $"); 47 #endif /* not lint */ 48 49 #include <sys/param.h> 50 51 #include <ctype.h> 52 #include <err.h> 53 #include <errno.h> 54 #include <fcntl.h> 55 #include <stdio.h> 56 #include <stdlib.h> 57 #include <string.h> 58 #include <unistd.h> 59 60 #define DEFLINE 1000 /* Default num lines per file. */ 61 62 static int file_open; /* If a file open. */ 63 static int ifd = -1, ofd = -1; /* Input/output file descriptors. */ 64 static char *fname; /* File name prefix. */ 65 static size_t sfxlen = 2; /* suffix length. */ 66 67 int main(int, char **); 68 static void newfile(void); 69 static void split1(off_t); 70 static void split2(off_t); 71 static void usage(void) __attribute__((__noreturn__)); 72 static size_t bigwrite(int, void const *, size_t); 73 74 int 75 main(int argc, char *argv[]) 76 { 77 int ch; 78 char *ep, *p; 79 off_t bytecnt = 0; /* Byte count to split on. */ 80 off_t numlines = 0; /* Line count to split on. */ 81 size_t namelen; 82 long name_max; 83 84 while ((ch = getopt(argc, argv, "0123456789b:l:a:")) != -1) 85 switch (ch) { 86 case '0': case '1': case '2': case '3': case '4': 87 case '5': case '6': case '7': case '8': case '9': 88 /* 89 * Undocumented kludge: split was originally designed 90 * to take a number after a dash. 91 */ 92 if (numlines == 0) { 93 p = argv[optind - 1]; 94 if (p[0] == '-' && p[1] == ch && !p[2]) 95 p++; 96 else 97 p = argv[optind] + 1; 98 numlines = strtoull(p, &ep, 10); 99 if (numlines == 0 || *ep != '\0') 100 errx(1, "%s: illegal line count.", p); 101 } 102 break; 103 case 'b': /* Byte count. */ 104 if (!isdigit((unsigned char)optarg[0]) || 105 (bytecnt = strtoull(optarg, &ep, 10)) == 0 || 106 (*ep != '\0' && *ep != 'k' && *ep != 'm')) 107 errx(1, "%s: illegal byte count.", optarg); 108 if (*ep == 'k') 109 bytecnt *= 1024; 110 else if (*ep == 'm') 111 bytecnt *= 1024 * 1024; 112 break; 113 case 'l': /* Line count. */ 114 if (numlines != 0) 115 usage(); 116 if (!isdigit((unsigned char)optarg[0]) || 117 (numlines = strtoull(optarg, &ep, 10)) == 0 || 118 *ep != '\0') 119 errx(1, "%s: illegal line count.", optarg); 120 break; 121 case 'a': /* Suffix length. */ 122 if (!isdigit((unsigned char)optarg[0]) || 123 (sfxlen = (size_t)strtoul(optarg, &ep, 10)) == 0 || 124 *ep != '\0') 125 errx(1, "%s: illegal suffix length.", optarg); 126 break; 127 default: 128 usage(); 129 } 130 argv += optind; 131 argc -= optind; 132 133 if (*argv != NULL) { 134 if (strcmp(*argv, "-") == 0) 135 ifd = STDIN_FILENO; 136 else if ((ifd = open(*argv, O_RDONLY, 0)) < 0) 137 err(1, "%s", *argv); 138 ++argv; 139 } 140 141 errno = 0; 142 if ((name_max = pathconf(".", _PC_NAME_MAX)) == -1 && 143 errno != 0) 144 err(EXIT_FAILURE, "pathconf"); 145 if (*argv != NULL) { 146 namelen = strlen(*argv) + sfxlen; 147 if (name_max != -1 && namelen > name_max) 148 errx(EXIT_FAILURE, "Output file name too long"); 149 if ((fname = malloc(namelen + 1)) == NULL) 150 err(EXIT_FAILURE, NULL); 151 (void)strcpy(fname, *argv++); /* File name prefix. */ 152 } else { 153 if (name_max != -1 && 1 + sfxlen > name_max) 154 errx(EXIT_FAILURE, "Output file name too long"); 155 if ((fname = malloc(sfxlen + 2)) == NULL) 156 err(EXIT_FAILURE, NULL); 157 fname[0] = '\0'; 158 } 159 160 if (*argv != NULL) 161 usage(); 162 163 if (numlines == 0) 164 numlines = DEFLINE; 165 else if (bytecnt) 166 usage(); 167 168 if (ifd == -1) /* Stdin by default. */ 169 ifd = 0; 170 171 if (bytecnt) 172 split1(bytecnt); 173 else 174 split2(numlines); 175 176 return 0; 177 } 178 179 /* 180 * split1 -- 181 * Split the input by bytes. 182 */ 183 static void 184 split1(off_t bytecnt) 185 { 186 off_t bcnt; 187 ssize_t dist, len; 188 char *C; 189 char bfr[MAXBSIZE]; 190 191 for (bcnt = 0;;) 192 switch (len = read(ifd, bfr, MAXBSIZE)) { 193 case 0: 194 exit(0); 195 /* NOTREACHED */ 196 case -1: 197 err(1, "read"); 198 /* NOTREACHED */ 199 default: 200 if (!file_open) { 201 newfile(); 202 file_open = 1; 203 } 204 if (bcnt + len >= bytecnt) { 205 /* LINTED: bytecnt - bcnt <= len */ 206 dist = bytecnt - bcnt; 207 if (bigwrite(ofd, bfr, dist) != dist) 208 err(1, "write"); 209 len -= dist; 210 for (C = bfr + dist; len >= bytecnt; 211 /* LINTED: bytecnt <= len */ 212 len -= bytecnt, C += bytecnt) { 213 newfile(); 214 /* LINTED: as above */ 215 if (bigwrite(ofd, 216 C, bytecnt) != bytecnt) 217 err(1, "write"); 218 } 219 if (len) { 220 newfile(); 221 /* LINTED: len >= 0 */ 222 if (bigwrite(ofd, C, len) != len) 223 err(1, "write"); 224 } else 225 file_open = 0; 226 bcnt = len; 227 } else { 228 bcnt += len; 229 /* LINTED: len >= 0 */ 230 if (bigwrite(ofd, bfr, len) != len) 231 err(1, "write"); 232 } 233 } 234 } 235 236 /* 237 * split2 -- 238 * Split the input by lines. 239 */ 240 static void 241 split2(off_t numlines) 242 { 243 off_t lcnt; 244 size_t bcnt; 245 ssize_t len; 246 char *Ce, *Cs; 247 char bfr[MAXBSIZE]; 248 249 for (lcnt = 0;;) 250 switch (len = read(ifd, bfr, MAXBSIZE)) { 251 case 0: 252 exit(0); 253 /* NOTREACHED */ 254 case -1: 255 err(1, "read"); 256 /* NOTREACHED */ 257 default: 258 if (!file_open) { 259 newfile(); 260 file_open = 1; 261 } 262 for (Cs = Ce = bfr; len--; Ce++) 263 if (*Ce == '\n' && ++lcnt == numlines) { 264 bcnt = Ce - Cs + 1; 265 if (bigwrite(ofd, Cs, bcnt) != bcnt) 266 err(1, "write"); 267 lcnt = 0; 268 Cs = Ce + 1; 269 if (len) 270 newfile(); 271 else 272 file_open = 0; 273 } 274 if (Cs < Ce) { 275 bcnt = Ce - Cs; 276 if (bigwrite(ofd, Cs, bcnt) != bcnt) 277 err(1, "write"); 278 } 279 } 280 } 281 282 /* 283 * newfile -- 284 * Open a new output file. 285 */ 286 static void 287 newfile(void) 288 { 289 static int fnum; 290 static int defname; 291 static char *fpnt; 292 int quot, i; 293 294 if (ofd == -1) { 295 if (fname[0] == '\0') { 296 fname[0] = 'x'; 297 fpnt = fname + 1; 298 defname = 1; 299 } else { 300 fpnt = fname + strlen(fname); 301 defname = 0; 302 } 303 ofd = fileno(stdout); 304 } 305 /* 306 * Hack to increase max files; original code wandered through 307 * magic characters. Maximum files is 3 * 26 * 26 == 2028 308 */ 309 fpnt[sfxlen] = '\0'; 310 quot = fnum; 311 for (i = sfxlen - 1; i >= 0; i--) { 312 fpnt[i] = quot % 26 + 'a'; 313 quot = quot / 26; 314 } 315 if (quot > 0) { 316 if (!defname || fname[0] == 'z') 317 errx(1, "too many files."); 318 ++fname[0]; 319 fnum = 0; 320 } 321 ++fnum; 322 if (!freopen(fname, "w", stdout)) 323 err(1, "%s", fname); 324 } 325 326 static size_t 327 bigwrite(int fd, const void *buf, size_t len) 328 { 329 const char *ptr = buf; 330 size_t sofar = 0; 331 ssize_t w; 332 333 while (len != 0) { 334 if ((w = write(fd, ptr, len)) == -1) 335 return sofar; 336 len -= w; 337 ptr += w; 338 sofar += w; 339 } 340 return sofar; 341 } 342 343 344 static void 345 usage(void) 346 { 347 (void)fprintf(stderr, 348 "Usage: %s [-b byte_count] [-l line_count] [-a suffix_length] " 349 "[file [prefix]]\n", getprogname()); 350 exit(1); 351 } 352
Indexes created Wed Oct 01 12:09:54 GMT 2025